Artificial Intelligence
Episodes whose main subject is AI itself: building with, training, evaluating, or debating AI systems. AI used only as a tool for another topic goes to that topic.
#5908: One Database Won't Fix Your Fragmented Stack
Rita has two employees, four apps, and one fear: which database wins when they disagree? Here's the field-by-field answer.
#5903: Video Tokens, Omni, and the Local VRAM Math
Video costs ~100 tokens per second. That single number explains why ChatGPT hides video upload and why local inference hits a KV cache wall.
#5900: Serving DeepSeek-V4.1-Flash on Your Own GPU
Five inference engines, one frontier MoE, and a shortlist that collapses depending on which GPU you already bought.
#5899: Same GPU, Same Weights, Different Answers
Identical hardware, identical weights, different engine — and answers that shift by up to 16.6 points.
#5898: ROCm 10.1: AMD's Software Stack Finally Grows Up
Three ROCm releases in six months — build system, agentic tooling, data movement. What actually changed for AMD GPU owners.
#5878: How a Sloth Learned to Talk: Corn's Origin Story
A neural interface gave Corn the bandwidth for speech. Nobody can explain where the accent, the manner, or Mongolia came from.
#5873: Abliteration: How $5 Undoes a Model's Guardrails
A rank-one weight edit costs under $5 and strips refusal behavior from open models. Then an automated tool beat human experts at it.
#5870: Homebox Fork's Two Backend Scripts Need a Real Job Runner
A Homebox fork has two incremental scripts — WebP conversion and AI image enrichment — and no scheduler built to run them safely.
#5869: Scraping AliExpress Specs: Selectors, Agents, or Neither?
Daniel wants to snapshot AliExpress spec sheets into his inventory system. The selector approach is already dead — here's what actually works.
#5868: Fine-Tuning Parakeet for Um-Free, Hebrish Speech
Parakeet runs great on a phone — but fine-tuning it happens on a GPU. Here's the actual recipe for dropping ums and catching Hebrew.
#5865: Four Pocket Players That Won't Sell You a Device Today
Mono, Fonix One, Sudo, and DuRoBo Krono all promise phone-free podcast listening — and not one of them is ready to ship today.
#5862: One Show, Eighteen Feeds: Apple's Duplicate Content Rules
One show, eighteen topic feeds. Apple's duplicate content rules have a specific answer — and it isn't the one you'd hope for.
#5861: Inside Cloudflare's Bot Check: TLS, JA4, and the Sand-Timer
What actually happens during those two seconds of "Checking your browser"? The verdict is made before a single byte of JavaScript runs.
#5860: Who Can Actually Train an LLM From Scratch?
Only a handful of organizations can truly train a model from random weights. Here's why nobody can count them.
#5859: Slow Labs, Fast Ships: What Release Cadence Actually Tells Us
Anthropic ships premium models every 22.5 days — the fastest lab in the industry. So much for the slow-and-careful story.
#5858: Routing AI Models: When 3 Rules Beat a Trained Router
A deep dive into AI model routing — and why a learned router may not beat three rules you wrote by hand.
#5857: Building a Scorecard for Our Own AI Writers
Daniel wants a real scorecard for the agent that writes this show — repetition, dialogue quality, and whether the retrieval harness actually works.
#5853: Who Actually Competes With Hugging Face?
2.96 million model repos, and 85.6% have under 200 downloads. What ModelScope, Kaggle, and Ollama actually do differently.
#5851: The Earbud Assistant Problem Nobody Solves
The AI model is solved. So why does the assistant in your ear still fail? The answer is the unglamorous plumbing nobody demos.
#5850: Ambient Computing: Why Glasses Beat the Phone
Weiser's 1991 vision said tech should disappear. Thirty-five years later, the device that finally worked wasn't new — it was glasses.